Re: A Question On Many-To-Many Linking Table(s)

From: Graham Johnson <johnson_at_NOSPAMpilot.msu.edu>
Date: Wed, 25 Sep 2002 22:22:02 GMT
Message-ID: <MPG.17fbe4f4f87a3d9a989683_at_netnews.attbi.com>


In article <c0d87ec0.0209221923.42b0561b_at_posting.google.com>, 71062.1056 _at_compuserve.com (--CELKO--) says...
> aleatory_at_hotmail.com (aleatory) wrote in message news:<a68a4ee0.0209220745.4442210e_at_posting.google.com>...
> > Hi all,
> >> According to some of the database books I have, they say that
> many-to-many implies a linking table(s). <<
>
> That is a terrible term; tables are either entities or relationships
> and not both. The term "link" is an old one we used for pointers, as
> in "linked list" and that low level concept has no place in a
> relational model.

The term "link" is also a not-so-old UML term for a (runtime) instance of an association. Link : Object :: Association : Class.

Searching on Google, I found the following different terms used to describe the table or entity created when resolving a many-to-many relationship (and there are certainly more):

association/associative entity/table
bridge/bridging entity/table
composite table/entity
intersect/intersecting/intersection entity/table join/joining entity/table
junction entity/table
link/linking entity/table
relationship entity/table

(Not to mention many descriptions of the process that don't give a name to the table at all).

In the unfortunate absence of a standard name, I think "linking table" is as valid a term as any other reasonably descriptive one. Received on Thu Sep 26 2002 - 00:22:02 CEST

Original text of this message